Bernardo B. Gatto
Federal University of Amazonas
36 Papers
122 Citations
Bernardo B. Gatto is an academic researcher from Federal University of Amazonas. The author has contributed to research in topics: Subspace topology & Discriminative model. The author has an hindex of 6, co-authored 32 publications. Previous affiliations of Bernardo B. Gatto include University of Tsukuba.
Chat about Author
Papers
Enhanced Grassmann discriminant analysis with randomized time warping for motion recognition
TL;DR: The validity of the proposed framework, RTW+eGDA, is demonstrated through experiments on motion recognition using the publicly available Cambridge gesture, KTH action, and UCF sports datasets.
21
Mutual singular spectrum analysis for bioacoustics classification
Bernardo B. Gatto,Juan Gabriel Colonna,Eulanda Miranda dos Santos,Eduardo F. Nakamura +3 more
- 01 Sep 2017
TL;DR: A novel bioacoustics signals classification method where no preprocessing techniques are involved and which is able to match sets of signals, which is independent of the signals length is proposed.
18
Discriminative canonical correlation analysis network for image classification
Bernardo B. Gatto,Eulanda Miranda dos Santos +1 more
- 01 Sep 2017
TL;DR: This work introduces a discriminative canonical correlation network (DCCNet), that employs filters constructed from discrim inative canonical correlations analysis (D CC), and demonstrates the applicability of DCCNet through experiments on four datasets.
15
Tensor analysis with n-mode generalized difference subspace
Bernardo B. Gatto,Eulanda Miranda dos Santos,Alessandro L. Koerich,Kazuhiro Fukui,Waldir S. S. Junior +4 more
TL;DR: In this article, the authors proposed a tensor representation for multi-dimensional data classification, which is based on generalized difference subspace (GDS) to reduce data redundancy and reveal discriminative structures.
15
•Posted Content
Tensor Analysis with n-Mode Generalized Difference Subspace
Bernardo B. Gatto,Eulanda Miranda dos Santos,Alessandro L. Koerich,Kazuhiro Fukui,Waldir S. S. Junior +4 more
TL;DR: The proposed approach outperforms methods commonly used in the literature without adopting pre-trained models or transfer learning and addresses the problem of representing and classifying tensor data for gesture and action recognition.